Has AV's time come?

[Lancer]

AV also relies too much on cute set plays. They worked before but everybody in the conference has seen them more than a couple of times. No changes to the program though.

As I said elsewhere, this team's confidence is in pieces and all chuckles seems to say is 'they will work themselves out of this...' I guess Vigneault has given up on trying to motivate them or get their heads back to a place where they play to win instead of waiting to fail.

He's been a very successful coach, and no doubt will do well wherever he goes next, but it's time to move on.

[Topper]

A very good piece by Cam Charron:

http://canucksarmy.com/2013/3/19/in-pra ... ose-models

LOL

Coles notes, the team may suck, but their smoke and mirror numbers are great.

[Vpete]

So then in the 2011 SCF Cam would say Vancouver just didn't have goaltending?

More importantly a lot of his ideas get really muddled in his over-reaching desire and quest to convince you that his way is the only way of looking at stats.

I've had a conversation with Gabe Desjardin- I asked him if he ever tried tracking what happens off of rebounds? Nope too hard but he said, "there's nothing there."

Gabe pulls his data from what is currently tracked by the NHL as does Greg Sinclair in his super shot analysis. http://www.somekindofninja.com As do all the others. Try tracking something during a game like zone entries. What makes anyone's definition of a zone entry more precise and accurate?

Where I struggle with all of this is that these 'stats' guys want to use the numbers to say what is a good hockey player or as predictors of future play. It all happens after the fact which makes sense but they rarely ever go out on a limb- GAbe did last year with his prediction the Wild would collapse on goaltending.

None of these guys said that about the Blues- oh wait. What about the Sens this year? So all they have been able to do is predict goaltending regression. Thomas Drance has done an admiarably job of predicting goal scoring totals for the CAncuks but that's about it.

The problem is these guy all want James' Mirtle's job, or Bochford's or whatever and there aren't that many around. The stats stuff gives them a purpose and a place outside of what is already there. Funny thing is, it's like the new trade rumours from 2005. Smoke and Mirrors as Topper said.

I'll look forward when to when NHL teams use this: http://www.grantland.com/story/_/id/9068903/the-toronto-raptors-sportvu-cameras-nba-analytical-revolution

[Topper]

What I found funny in the links Waffle posted the other day is that some of these folks are now realizing their numbers don't work so the are manipulating them with more made up numbers failing to realize that errors are compounded in calculation.

It comes down to realizing your black box doesn't work so you use a smoke and mirrors to try and fix it.

Some of these folks math is absolutely horrendous. Cam stands out in this regard.

I wish the few who use these numbers as a predictor would use them in the correct way, the most common thing I see is they predict to an average rather than to a trend line. That is either a poor understanding of stats or flat out laziness.

My problems with Corsi and Fenwick are mainly that they both fall apart at the edges of data and that they rank all shot equally.

http://blogs.thescore.com/nhl/2013/03/12/why-advanced-statistics-approaches-hockey-completely-wrong-but-still-gets-things-right/

How about the 1972 Summit Series as an "aha" moment when we all realized quality shots are more important than shots.

The reason these advance stats guys don't use scoring chances is because it takes much longer to generate a meaningful data population and because the population is lower, variability if higher.

I have to keep going back to the ground breaking work of Bill James showing a direct correlation between minor league stats and pro stats for baseball. When a metric for hockey does the same.........oh boy.

[Vpete]

Holy shit- I just read the comments underneath and the same guys who are proponents of Corsi are saying it's a bad idea to shoot because it gives up possession. So why do you track and action that gives up possession?

Anywhooooo....

Don't even get me started on 'scoring' chances. It's a shit-show of complete and total ambiguity.

http://blogs.thescore.com/nhl/2013/03/12/why-advanced-statistics-approaches-hockey-completely-wrong-but-still-gets-things-right/

This is what Cam Charron told me was the definition of a scoring chance. Go scratch your head around that for a bit.

So what is a chance? Any time the puck enters that area? Or any time a shot is made from that area?

Then consider shot quality. The stats guys don't want to touch that because it doesn't mean anything. I asked David Johnson the same thing about shot quality- is a good shot for the skater not a poor shot for the goalie?

Silence

Simply put I want to know if my keeper is letting in goals scored by quality chances for the skater or if he has trouble on poor chances for the skater. Because if my goalie is getting hammered I can adjust the defense or I can then work on the trouble spots with the goalie. However, the scoring chances being tracked are more on the 'possession' determines everything slant and I just can't go that far.

It's funny in basketball rebounds are counted in terms of defensive and offensive possession- wouldn't it make sense to do the same in hockey?

[Topper]

Thanks Pete.

Following a link in the comments I got to the Super Shot page for Kenendy.
http://somekindofninja.com/nhl/index.php?season=Regular&year=2011-2012&shots=For&team=Pittsburgh+Penguins&ice_player_name=Tyler+Kennedy&withPlayer=On+Ice&player_name=&goalie_name=&event=Shots+and+Goals&game=Home+and+Away&strength=Even&time=Regulation&search=Search
This is approaching what I have been saying. Now, add in missed and blocked shots (not including them is akin to only charting pitches in the strike zone at the exclusion of balls) and then have a ranking for shot quality. Contour that data and you really have something.

I know the shot quality data is not available, but it could be tracked by a teams scouting staff.

Now do this for college, major junior and the AHL and you definitely see what is happening with prospects. That is where the real value lies.

[Waffle]

Thanks for the links. The articles and comments are very worthwhile reading.

[Topper]

The milking of these numbers is exactly what is implied by POD's comments about his self named stat and the Oilers.

[Vpete]

At least Mirtle is getting to the nuts and bolts of the analytics thing.

http://www.theglobeandmail.com/sports/hockey/moneyball-ization-of-hockey-pays-off-for-penguins/article10122427/

It shows that VOILA- looking at stuff in detail and tracking things based on accurate data designed for specific conclusions can work. You're just not going to get that stuff from NHL stats sheets pulled into Excel

[RoyalDude]

Does AV get Coach of the Year nods this season for managing a winning record with a ramshackle, stale line-up that his useless superior - Mike 'Snoozey' Gillis put in place for him?

[BingoTough]

Does AV get Coach of the Year nods this season for managing a winning record with a ramshackle, stale line-up that his useless superior - Mike 'Snoozey' Gillis put in place for him?

If it looks like a troll and posts like a troll...

You are seriously boring.

[RoyalDude]

Does AV get Coach of the Year nods this season for managing a winning record with a ramshackle, stale line-up that his useless superior - Mike 'Snoozey' Gillis put in place for him?

If it looks like a troll and posts like a troll...

Ad Hominem Alert!

[Tiger]

hmm so we blame the coach for the lousy production of the PP?? ( thought that was Newell Browns job? ) .. or
maybe we blame him for the lousy production of the #1 line?? ( Sedin's production is declining maybe because of aging and maybe because they are "One Trick Ponies" and unable to change the way they play? )
So lets examine with something other than rose collored glasses exactly what AV has to work with ( Yeah the Dude has a point )..
a #1 line that is producing at about 1/2 what it used to...
No #2 line.. Booth, Kesler both out..
So our goal production is coming from what are basically 2 #3 lines..
Our wins are dependent on winning low scoring games by playing great defensively and excellent goaltending..
AV has been coaching very well considering the talent on his team

[RoyalDude]

hmm so we blame the coach for the lousy production of the PP?? ( thought that was Newell Browns job? ) .. or
maybe we blame him for the lousy production of the #1 line?? ( Sedin's production is declining maybe because of aging and maybe because they are "One Trick Ponies" and unable to change the way they play? )
So lets examine with something other than rose collored glasses exactly what AV has to work with ( Yeah the Dude has a point )..
a #1 line that is producing at about 1/2 what it used to...
No #2 line.. Booth, Kesler both out..
So our goal production is coming from what are basically 2 #3 lines..
Our wins are dependent on winning low scoring games by playing great defensively and excellent goaltending..
AV has been coaching very well considering the talent on his team

Easy Big Fella, this post might put you in Troll category.

[Jovocop]

hmm so we blame the coach for the lousy production of the PP?? ( thought that was Newell Browns job? ) ..

Isn't Newell Brown part of the coaching team? AV is in charge of the overall coaching. He could force Brown to change the PP units if things don't work out. With Edler out, Garrison has performed well in the first power play unit. With Edler back on Tuesday, I am almost 100% sure that Edler will be back on the first unit even though he is not performing.